Rare find

Autoresearch. An AI assistant that improves your model overnight by running experiments on its own.

github.com/karpathy/autoresearch

Problem

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Audience

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Market

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Pricing

The full read on this project is a sign-in away — what it solves, who it serves, and where it sits in the market.

Vaya's read on this project

Problem, audience, market, and the verdict — sign in to see it.

Updates

March 2026

Merge pull request #342 from kaizen-38/feat/bug-fix Mar 26
Enhance README with more project context and links Mar 21
fix(analysis): define best_bpb before y-axis scaling Mar 19
Merge pull request #301 from indianspeedster/master Mar 16
add AMD ROCm fork to notable forks section Mar 16
Guard against infinite loop when no training shards exist, fix README… Mar 11
fix NaN loss not caught by fast-fail check Mar 11
fix(train): make NaN fast-fail check explicit Mar 11
Guard against infinite loop when no training shards exist, fix README… Mar 11
Include beginner's guide to neural networks Mar 9
fix NaN loss not caught by fast-fail check Mar 9
clarify that results.tsv should not be committed, leave untracked Mar 9
some docs on what to play with to make autoresearch better on smaller… Mar 9
reshuffle readme a bit and link to tiny stories for apple silicon gui… Mar 8
clarification to baseline run instruction, there was some language fr… Mar 8
add notable fork mlx Mar 8
Honor --download-workers instead of hardcoding 8 download workers Mar 7
add macos fork Mar 7
fix(prepare): honor --download-workers Mar 7
instruct the agent to also read README, should be good context Mar 7